Lay summary
The understanding of various processes on an atomistic level is important in many fields of science such as nano science, materials science or molecular biology. In an atomistic computer simulation one can calculate the positions of individual atoms and follow their motion.
Simulations are therefore an ideal tool to obtain an atomistic understanding.
We work on the improvement of methods for such atomistic simulations.
We want to make them faster and more accurate. The speed of a simulation is determined by the speed of the computer and the algorithm that is used for the simulation.
Atomistic simulations require typically a lot of computer time. Some of them run for several months on huge parallel computers.
We develop new and faster algorithms and in particular algorithms which will not slow down too much when the number of atoms in the simulation increases.
We also work on eliminating some approximations that are frequently made in such simulations.
In this way the accuracy of the simulation increases and it has more predictive power.
We apply the methods we have developed currently to problems in cluster physics, nano science and molecular biology.